The practice of chemistry rests upon three pillars- Theory, Synthesis, and Analysis.  To bring a chemical product into the market place efficiently, a program of development must evolve that rests upon the three pillars.

Ostensibly, in order to rationally synthesize- that is, conceive of a new substance and design a means to bring it into being- it is advantageous to have some kind of theoretical background in order to take advantage of the orbital formalisms of bond making and bond breaking.  It is certainly possible to do chemical synthesis without even a clear notion of atomic theory.  William Perkin was able to embark on a synthesis of quinine and (end up with a synthesis of Mauve Dye) in 1856 without the benefit of molecular orbital formalism. Of course, if he had the formalisms and the analytical technology, he might have actually come up with quinine and would henceforth been known as the father of pharmaceuticals rather than the father of the synthetic dye industry.

I have been witness to numerous product development cycles in the fabulous commercial world of specialty chemicals. If there has been one underlying theme to all of this product development that I have been a part of, it is that synthesis development is typically years ahead of the analytical devlopment.  Allow me to elaborate. 

[Please note that I am not talking about pharmaceutical product development.  I do not operate in that strange universe and I do not pretend to understand it or even desire to be in it.]

Somewhere a company with lots of R&D money to spend and dash of vision will arrive at a stage-gate in its new product development.  A collection of compounds will be identified as having solid potential for use in a profit making chemical enterprise.  Project managers will have to decide on a molecule to launch the project.  The molecule can be a final product with a specific identity, or it can be a substance used to facilitate a technology platform.

The drivers of the project will invariably be synthesis chemists and engineers.  They may choose to make the molecule of interest in-house.  If the molecule or material is the product to be sold, they will almost aways make it in house to capture the economies of vertical integration and scale.  But if the molecule of interest is a reagent, catalyst, initiator, or specialized intermediate requiring some black art, the developers may choose to farm out the molecule. 

In the latter case of reagent, intermediate, etc., farming out the molecule to a specialist vendor requires that the company disclose the identity of the species and probably a synthetic pathway.  Like dogs sniffing one another, a customer and vendor will circle around each other for a short while trying to assess the merits of the relationship.  Once an agreement to move past the disclosure stage is agreed upon, the vendor will set upon the task of noodling out a process. 

I believe it is axiomatic that analytical culture is different from synthetic culture.  Analytikkers live in a world of validation, significant figures, calibration curves, error analysis, and standard test methods.  Synthetikkers live in a world of space yields, solvent effects, reagents, exotherms, hazmats, filtration, distillation, etc.  Each group looks at product development from a different angle and imperative.

Here is the point I wish to make.  Compounds that have been recently discovered and submitted for scale-up are very often “new species”. That is, molecules that are not fully understood in terms of stability, contaminant profile, and importantly, analytical signature.  It would be best to take the time to fully investigate the compound. But to fill out the data table on a species that may not actually go forward is to commit precious time in a very risky way.  Usually, it seems, a candidate for process development is minimally characterized and put on a frantically short timeline for commercialization. 

Another axiom: If there is a hole, someone will fall in it.  Scale-up is often the beginning of the period I refer to as FMD, or “Failure Mode Discovery”.  During this FMD period up to and including pilot scale processing, it invariably transpires that in-process checks and analyses of intermediates is complicated by the improper choice of analytical method and failure to characterize side products. 

In their frenzy to meet deadlines and goals, synthetikkers may not be able to complete a crucial aspect of their job.  That would be to form a complete understanding of the process.  It includes the identification of side products and the fullest characterization of the product as possible.  It is crucial to find in-process markers that indicate that a reaction is proceeding swimmingly or that it is going afoul.  I believe it is squarely the responsibility of the synthesis chemist to survey the composition of critical intermediates and the final product mixture. 

While the preceeding seems obvious and even pedantic, the cost pressures on new product development are often severe and accordingly, processes are rushed out of R&D without much attention to the analytical issues.  I have seen new products from some of the world’s greatest R&D groups hit with severe quality issues in commercialization because analysts weren’t brought in to help with the characterization.

Analysts frequently need input with the development of quality control test methods for new substrates.  This is where the synthetikker can provide the crucial input.  Synthetic chemists must be well versed in the Three Pillars of Chemistry.  We acquire a theoretical background to support our synthetic activity, but we have the critical responsibility of knowing a variety of analytical techniques to validate our assertions that we have made a particular molecule.

Many times in our haste to get a project wrapped up, we rely on NMR for primary analytical data. Very often, NMR is perfectly satisfactory as a stand alone spec, as long as you do not need reliable data below 0.1 %. 

But NMR doesn’t always tell the whole story.  In fact, I have often seen fellow chemists throw up their hands in a gesture of complete frustration and give up when NMR fails to afford a clue to a process or product problem.  Basically, NMR is fast and affords structural details that are unavailable any other way.  Everything else is a science project.  

Having served in business development and product management, I can testify that unforseen quality issues can become show stoppers. It is not unusual to spend as much R&D time trying to noodle out unanticipated quality issues as it took to develop the product in the first place.

It is good to have two or three ways to quantitate purity. I’ve found it useful to have a good relationship with the analytikkers- one that allows for brainstorming and problem solving.